PART TWO: POWER AND ENGINEERING 312 Figure 5.13 (a) The air-standard Diesel (modified Carnot cycle) plotted on coordinates of pressure (P) and volume (V). The true Carnot cycle (1'), (2'), (3'), (1') was modified by Diesel as follows: (i) expansion was terminated at point (5) in order to provide a cylinder of reasonable dimensions (shown solid in the lower diagram, with the dashed lines showing the cylinder dimensions required by the true Carnot cycle); (ii) the constant temperature heat addition process (3')-(4) of the true Carnot cycle was replaced by the constant pressure heat addition process (3)-(4) in engines built after 1893 when it was found in the prototype to be impossible to produce a constant temperature combustion process. STEAM AND INTERNAL COMBUSTION ENGINES 313 (b) The air-standard compression ignition cycle plotted on coordinates of pressure (P) and volume (V). The heat addition (combustion in the real engine) processes tend to be different depending on the method of fuel injection. Research by Ricardo showed that fuel injected by air is much more finely divided than that resulting from solid injection, so burning is very fast and the rate of combustion can be controlled by the rate of fuel admission. Consequently, the pressure-volume diagram follows the path (2)–(3). With solid injection there is a delay before combustion starts, this is followed by a rapid rise in pressure (2')–(3'). At point (3') the pressure and temperature of the cylinder contents are so high that the remaining fuel ignites on entry, and, as with air injection, the rate of burning depends on the rate of fuel supply (3')–(4'). divided spray, at the end of the compression process, and is ignited by contact with the high temperature compressed air in the cylinder. Diesel’s original proposal, made in about 1890, was for an engine working on the Carnot cycle (see Appendix, p. 342), with a compression ratio of 50:1, a maximum pressure of 253bar (3675psi) and temperature of 800°C (1470°F), giving an air standard efficiency of 0.73 (Figure 5.13 (a)). However, the mean effective pressure of the proposed cycle was very small, which meant that a slight deviation from the design conditions would prevent the engine from operating because it was unable to overcome the internal friction. In 1892 the Maschinenfabrik Augsburg (later the Maschinenfabrik Augsburg-Nürnberg or MAN) contracted with Diesel to construct an experimental engine with a 15:1 design compression ratio in exchange for manufacturing rights. This engine was tested between July and September 1893. It was not able to run under its own power because the friction power exceeded the engine power, but it did demonstrate firstly that a charge of fuel could be ignited by compression alone; and, secondly, that when the fuel is injected it must be broken up into minute droplets, and that this was best achieved (at that time) if the fuel was sprayed into the cylinder by compressed air. In spite of the problems, the engine was sufficiently promising to attract a number of licensing agreements in 1893. In February 1897 a fourth experimental engine (see Figure 5.13 (a)) with a compression ratio of 11:1 was built, which, when tested, had a brake thermal efficiency of 30.7 per cent (fuel consumption 0.44lb mass/hphr; 0.27kg/kWhr), which was substantially higher than any contemporary heat engine. Further licences were taken up, but the move was premature because no licensee was able to build a reliable engine. Consequently, between 1897 and 1902 a major development programme was carried out by MAN to produce a satisfactory engine. During the next ten years or so, the Diesel engine gradually found its place in the stationary engine field with engines of 15–75kW (20–100hp) that were more convenient and more economical than gas engines. However, mobile application did not appear practical because of the bulky air compressor PART TWO: POWER AND ENGINEERING 314 required to produce the fuel injection ‘air blast’ and because of the low engine speed (piston speed: 3m/s or 590ft/min). The modern descendant of the Diesel engine is better called a compression ignition engine because it does not operate on Diesel’s theoretical cycle. Furthermore, beginning in the 1920s, when solid fuel injection was introduced (see p. 322) the mode of operation differed even more from the original Diesel concept (see Figure 5.13 (b)). The gas engine: 1880–1986 The history of the gas engine from 1880 onwards has two phases, the first lasting until about 1914, and the second from about 1925 to the present. Until 1914 the history of the gas engine is very closely tied to the available fuel. In its earliest days (1860 to 1880) coal-gas was used (LHV: 19,371kJ/m 3 or 520BTU/ft 3 ), but it was expensive and restricted the engines to power outputs of no more than 55kW (75hp). In 1878, J.E.Dowson invented the suction gas producer which generated a gas mixture (LHV: 5029kJ/m 3 or 135BTU/ft 3 ) by the partial combustion of coal with air and steam, and costing about one-tenth the price of coal-gas. Because its lower cost more than compensated for its smaller heating value compared to coal-gas, the use of the gas engine increased, along with its size (up to 75kW/100hp). In 1895, B.H.Thwaite demonstrated at the Glasgow Iron Works that blast furnace gas (approximate composition CO 35%, N 65%; and LHV 3725kJ/m 3 or 100BTU/ft 3 ) was a suitable engine fuel, and since it was otherwise a waste product of the iron smelting process, its small heating value was no disadvantage. The availability of large quantities of cheap gas and the need to drive big blast furnace blowers, resulted in the construction of extremely large gas engines for use in iron and steel works. One of the largest was a four- stroke cycle engine of 7500kW (10,000hp) that was installed at the South Chicago Works of the Illinois Steel Company in 1931. Nevertheless, by about 1914 it was clear that the gas engine could not compete with the compression ignition engine or the steam turbine. It was the commercial exploitation of natural gas (mostly methane, with smaller quantities of other hydrocarbons; LHV: 39,113kJ/m 3 or 1050BTU/ft 3 ) in the United States, starting in the mid-1920s, that revived the fortunes of the gas engine. This gas has to be conveyed by pipeline from its source to its point of use, requiring compressors located at intervals. These can be driven by reciprocating engines (sometimes gas turbines) burning as fuel the gas being pumped. The availability of well-developed compression ignition engines, together with a fuel in the pipeline at a high pressure, suggested that these engines could be adapted to burn natural gas by replacing the compression ignition cylinder head by one incorporating a fuel injector, and either a spark STEAM AND INTERNAL COMBUSTION ENGINES 315 plug or an oil fuel injector to provide a pilot flame, as an alternative to the electric spark. Modern forms of this engine operate with compression ratios as high as 13:1. Supercharging: 1909–1930 Supercharging describes any technique intended to increase the density of the air (oxygen) supplied to an internal combustion engine. It can be accomplished by either raising its pressure by a compressor (‘blower’) or decreasing its temperature by the addition of volatile liquids; however the name is usually associated with the first method. So-called charge-air cooling has also been considered for compression ignition engines. Supercharging was apparently first proposed by Daimler, but was not tried until 1878 when Dugald Clerk employed it on his two-stroke gas engine (see p. 307). The first significant use of supercharging was in aircraft engines, which operate at a low ambient pressure. This was first proposed by Auguste Rateau in 1914, and the idea was taken up about 1918 by, among others, S.A.Moss in the United States and A.Büchi in Switzerland. The supercharger has also been used extensively in racing cars, since its first application by Mercedes in 1922, to increase the power output by an engine of a given size. High power compression ignition engines have adopted supercharging so extensively, in order to increase the power-to-weight ratio, that supercharged engines of this type are the rule rather than the exception. Supercharging can be provided by piston pumps, vane pumps, Roots compressors and the centrifugal compressor. They can be driven mechanically off the engine crankshaft, or, where space is not limited, electric motors and even steam turbines have been used. However, it is the gas turbine, as originally proposed by Rateau, using the exhaust gases from the engine, and driving a centrifugal compressor that has been the most significant development in supercharging (nowadays commonly called turbocharging). The application of the gas turbine driven supercharger is not simple. Firstly, if the engine is to produce power, the gas turbine-compressor combination must have a high efficiency (more than 55 per cent) and the temperature of the exhaust gas entering the gas turbine must be greater than about 400°C (750°F). Secondly, difficulties can arise because of the overlap of the opening of the intake and exhaust valves, which is required to minimize the effects of valve and gas inertia and is, furthermore, required in the two-stroke engine to provide adequate scavenging. If, when this situation occurs the pressure in the exhaust manifold is higher than the intake pressure, then back flow can occur. This situation can be overcome in two ways. In constant pressure supercharging the volume of the exhaust system can be made large enough to keep its pressure essentially constant and never exceeding that in the inlet manifold. This method, invented by Rateau in 1914, is used in aircraft engines. In pulse supercharging the exhaust PART TWO: POWER AND ENGINEERING 316 manifold is divided so that cylinders exhausting into a particular manifold do not produce interfering pressure pulses. Büchi devised this method in about 1925 as a result of tests that he started in 1909. The reasons for choosing a particular supercharging technique are complex, but in general pulse supercharging is preferred in two-stroke cycle engines, because it makes better use of the energy in the exhaust gas, thereby compensating for the decrease in the gap temperature caused by the excess air necessary to ensure efficient cylinder scavenging. Pulse supercharging involves complicated exhaust manifold arrangements, so the constant pressure method is desirable where first costs must be minimized, as in automobile applications. The spark ignition engine The spark ignition engine is in almost universal use for automobiles, and it is also used as an aircraft engine when the power and speed capabilities of the gas-turbine are not required. The modern spark ignition engine has been subjected in its lifetime to some of the most intensive scientific study of any thermal prime mover because of its sensitivity to the fuel used, and because it is a potent and very widespread source of atmospheric pollutants. 1900–1920: The high speed spark ignition engine The need to produce a simple and reliable automobile engine led to changes in combustion chamber shape, and increase in the number of cylinders (from 1900 to 1915), improved ignition systems, decreasing gasoline volatility, better fuel supply systems, and forced lubrication (1908). Combustion chambers were continually changing in order to provide an engine that was simple to manufacture and maintain, had satisfactory power output and fuel economy, and avoided detonation (see below). The most widely used combustion chamber was the side-valve or L-head, which, however, had a strong tendency to produce detonation. Research conducted by H.R.Ricardo between 1912 and 1918 showed that the compression ratio at which detonation first occurred could be raised from 4.0 to 4.8 if the turbulence of the air-fuel charge was increased and the spark plug was located nearer the centre of the cylinder. 1920–1945: detonation This particular phenomenon produces an unpleasant noise, but more importantly, it indirectly has a deleterious effect on the fuel consumption, because it limits the maximum useable engine compression ratio. STEAM AND INTERNAL COMBUSTION ENGINES 317 Detonation (‘knock’) was first identified and differentiated from pre-ignition (due to hot surfaces) in 1906 by Dugald Clerk and H.R.Ricardo. It is defined as self-ignition of the end gas (the unburnt portion of the charge) as a consequence of its temperature and pressure having been increased by compression resulting from the expansion of the burning portion of the charge. This produces high frequency oscillations in pressure, and hence the characteristic engine noise. Detonation first became a significant consideration in 1905 as more volatile petrol came into use, but it was not until the First World War that it became critical as a result of its effect on aircraft engine performance. Consequently, a group in the United States, under the direction of C.F.Kettering of the Dayton Metal Products Company, developed a special aviation fuel consisting of a mixture of cyclohexane and benzene which allowed aircraft engines to operate without detonation at compression ratios up to 8:1. Following the war, work on solving the detonation problem adopted two approaches: firstly, the search for a chemical additive (dope) which would suppress detonation, was undertaken by Kettering, T.A.Boyd, and Thomas Midgley, all of General Motors. The second approach, involving the best combination of engine design parameters and fuel to minimize detonation, was initiated by Ricardo. Boyd, in 1919, starting from his discovery in 1916 that iodine could suppress detonation, found that aniline was also suitable. However, neither compound was entirely practical (iodine was corrosive and aniline had a bad smell). Eventually in 1921, as a result of a lengthy search, lead tetraethyl was identified as a suitable detonation suppressant. This historic discovery made the modern high compression spark ignition engine possible (see Figure 5.14). In England, Ricardo formed his own company in 1917 with the objective of finding the combination of fuel and combustion chamber form that would eliminate detonation. As a result he decided that the detonation tendencies of engines and of fuels must be defined in terms of standard fuels and standard engines. In 1924 he introduced a mixture of toluene (resists detonation) and pure heptane (readily detonates). The detonation characteristics of a candidate fuel could then be expressed in terms of the percentage of toluene in the standard fuel that matches the detonation characteristics of the given fuel. Toluene was later (1927) replaced by octane, hence the now-familiar octane number. Since detonation depends on the engine compression ratio, as well as the fuel, the candidate fuel must be tested in a standard engine in which the compression ratio may be varied while the engine is running. Ricardo was first to build such an engine (in 1919). Eventually in 1931 a committee (the Cooperative Fuel Research Committee) of interested parties (engine manufacturers, oil refiners) in the United States, under the leadership of H.L.Horning of the Waukesha Engine Company, produced a standard engine PART TWO: POWER AND ENGINEERING 318 design, known today as the CFR Fuel Research Engine, which is now a world- wide standard for detonation research and testing. 1945–1986: power, efficiency and cleanliness Initially, after the war ended in 1945, engines of ever increasing power, and mean effective pressure were produced (Figure 5.14). There was also a progressive increase in the compression ratio (Figure 5.14) and the octane number of available petrols. This trend was arrested by the combined effects of environmental concerns (since 1963) and the ‘energy crisis’ (1973). Scientific investigations, particularly in southern California, in the late 1940s and early 1950s had demonstrated that hydrocarbons, carbon monoxide, and various oxides of nitrogen, produced by automobile spark ignition engines, were important air pollutants. It was clear that with the anticipated increase in the number of motor vehicles, particularly in large urban areas, control of emissions from their engines was essential. The Figure 5.14: Historical trend of the performance parameters of four-stroke cycle spark ignition automotive engines 1920–85. Unfortunately, the data available to the author were limited as shown. The curves are based on data in C.F.Taylor The Internal Combustion Engine in Theory and Practice, vol. 2, revised edition (The MIT Press, Cambridge, Mass., 1985) and Automotive Industries, vol. 165, no. 6 (1985), p. 521. STEAM AND INTERNAL COMBUSTION ENGINES 319 government of the State of California, and then the United States government, enacted legislation on automobile emissions, and comparable regulations were imposed in many countries outside the United States. The environmental regulations imposed by the US government on the automobile engine were embodied in the so-called 1963 Clean Air Act, amended in 1968, 1970 and 1977. The practical realization of the various techniques devised to meet the legal requirements has resulted in the need to solve some of the most challenging technical problems presented by the internal combustion engine. There are five main emission control methods, (a) Air injection into the exhaust manifold to burn hydrocarbons and carbon dioxide, (b) Exhaust gas recirculation (EGR), which lowers nitrogen oxides emitted by the engine by diluting the intake fuel-air mixture, thus decreasing the maximum cylinder gas temperature, (c) Oxidizing catalytic converter, which is located in the exhaust line and assists the oxidation of hydrocarbons and carbon dioxide. In 1981 a so-called three-way catalytic converter, which additionally converts oxides of nitrogen to nitrogen, was introduced. This has allowed the engine to operate at conditions suitable for minimum fuel consumption, (d) Electro-mechanical carburettor which controls the engine fuel supply by sensing the oxygen level in the exhaust gas before it enters the three-way catalytic converter. The sensor and the carburettor are linked through a microcomputer to provide a control system that maintains the air-fuel ratio at 14.6±0.2. (e) Fuel injection into the inlet manifold or at the cylinder intake ports, which allows even more precise control of the air-fuel ratio than the electro-mechanical carburettor. The effort to meet emission standards in mass-produced engines has resulted in a complicated engine that has a higher first cost and is expensive to maintain. It is possible that this situation could be avoided if the combustion processes in the engine cylinder could be fundamentally changed. It has been known for some time that a marked reduction in carbon monoxide and oxides of nitrogen emissions (but not hydrocarbons) can be obtained if the fuel-air mixture was weaker than that required by purely chemical considerations (the so-called stoichiometric mixture): however, the engine has a low efficiency and runs very roughly. This can be avoided by using a stratified charge (see p. 306–7), and a substantial effort has been made by automobile manufacturers to develop an engine of this type. Because its operation is sensitive to variations in design and conditions of use, it has been found difficult to employ this type of engine in a mass-produced vehicle: the only commercially available engine in the late 1980s was the Honda Compound Vortex Controlled Combustion (CVCC) engine. Aircraft engines The contribution of the aircraft engine to the development of the internal combustion engine has been particularly important in producing engines of PART TWO: POWER AND ENGINEERING 320 low specific weight (kg/kW). Many of the features that have contributed to this development have reappeared in the 1980s in low specific weight automobile engines with minimum fuel consumption and pollution. The automobile engines available to the earliest aircraft builders (c.1900) were too heavy and of too low power, and the first practical engines produced by the Wright brothers (see p. 622) and by C.M.Manly for Langley’s ‘Aerodrome’ of 1903, were designed specifically for use in aeroplanes. Manly’s engine was remarkable in having a specific weight of 1.45kg/kW (2.38lb mass/ hp), which was not improved on until the American ‘Liberty’ engine of 1918. The Manly engine was unique in its anticipation of many features that became standard on later aircraft engines: a radial arrangement of the cylinders with a master connecting rod, the cam and valve gear arrangement, and crankcase, cylinders and other parts that were machined all over to carefully controlled dimensions. Aircraft engines following the initial period (c.1903–14) were of two main types. The air-cooled radial engine was based on the Gnome rotary (cylinders rotated around the crank shaft) radial engine, which was the most popular aircraft engine up to the First World War and was used by both sides in the conflict. It was mostly, but not exclusively, used by civilian aircraft operators, because of its simpler cooling arrangements. The first large radial air-cooled engine of modern design was the Pratt and Whitney ‘Wasp’ (1927). It employed a mechanically driven centrifugal supercharger, as well as a forged and machined aluminium crankcase and cylinder head. The liquid-cooled vee engine had its origins (c.1915) in the Hispano-Suiza engine (1.9kg/kW or 3.1lb mass/hp). The basic structure consisted of a cast aluminium crankcase with en bloc water jackets. Initially, aircraft engines used water for cooling, but this was changed c.1932 by the Curtiss Company in the US when they introduced a water-ethylene glycol mixture. This boiled above the boiling point of water, so that the higher temperature difference between the coolant and the ambient air allowed the use of a smaller, lower drag radiator. The Curtiss Company, starting in 1920, developed a succession of in- line engines, based on the Hispano-Suiza engine, which had considerable success in racing, particularly in the Schneider Trophy sea-plane races (see p. 630). This led the Rolls-Royce Company in Britain to design the ‘Kestrel’ engine in 1927 (1.1kg/kW or 1.77lb mass/hp), followed by the ‘R’ (for racing) type, and then the ‘Merlin’ (0.78kg/kW or 1.28lb mass/hp). This last engine, together with the Wright ‘Cyclone’ and Wright ‘Wasp’ radial engines, represent the pinnacle of the reciprocating spark ignition aircraft engine development. Figure 5.15 summarizes the development of the aircraft piston engine between 1900 and 1960. STEAM AND INTERNAL COMBUSTION ENGINES 321 The compression ignition engine The modern compression ignition engine is the most efficient thermal prime mover available today, and the pollutants in the exhaust are less than those produced by spark ignition engines. It is the engine of choice in non-aircraft transportation applications where low operating costs are more important than the first cost. However, the compression ignition engine was not readily adaptable to transport applications until about 1930 when engine speeds increased and solid injection replaced air-blast fuel injection. Figure 5.15: Historical trend of the performance parameters of aircraft spark- ignition engines 1900–60. Adapted with permission from: C.F.Taylor ‘Aircraft Propulsion: A Review of the Evolution of Aircraft Power Plants’, Smithsonian Report for 1962 (Smithsonian Institution, Washington, D.C., 1962). . type are the rule rather than the exception. Supercharging can be provided by piston pumps, vane pumps, Roots compressors and the centrifugal compressor. They can be driven mechanically off the engine. greater than about 400°C (750°F). Secondly, difficulties can arise because of the overlap of the opening of the intake and exhaust valves, which is required to minimize the effects of valve and gas inertia. (b)). The gas engine: 1880–1986 The history of the gas engine from 1880 onwards has two phases, the first lasting until about 1914, and the second from about 1925 to the present. Until 1914 the history